feat(hosts): surface per-host connectivity on /hosts; drop the unwritable table (#521)
`IHostConnectivityProbe` was a dead surface: eleven drivers implement it, `GetHostStatuses()` had ZERO production call sites, and `OnHostStatusChanged` had no subscriber outside the Galaxy driver's own aggregator. Per-host connectivity was computed by every driver and read by nobody. The issue offered "build the publisher or delete it". The publisher as its entity doc described it — driver nodes upserting `DriverHostStatus` rows — is not buildable: per-cluster mesh Phase 4 gates `AddOtOpcUaConfigDb` on the `admin` role, so a driver-only node has no ConfigDb connection to write rows with. So the capability is kept and the transport changed. `DriverHealthChanged.HostStatuses` now carries the probe result to `/hosts` as a Hosts column. That channel already reached the page, already survives the mesh split via the Phase 5 gRPC telemetry stream, and already replays a last-value snapshot on re-subscribe — so per-host state re-primes after a reconnect without a durable store. Both halves of the interface finally do what they are for: the event triggers a prompt publish, the pull is the source of truth. The point of the column is the case the driver-level state chip structurally cannot express: a multi-device driver stays aggregate-Healthy while ONE of its devices is unreachable. Two traps, both pinned by tests that were falsified against the prod code: - The host digest MUST be in the publish fingerprint. On a single-host-down transition every other fingerprint component is unchanged, so the dedup would swallow exactly the publish carrying the news — the trap that already bit the rediscovery signal. Removing it turns the guard test red, verified. - null (no probe) must stay distinct from empty (probe with no hosts). proto3 cannot tell an absent repeated field from an empty one, hence the explicit `has_host_statuses` flag; collapsing them would render every probe-less driver as one whose devices are all fine. Dropped: the DriverHostStatus entity, enum, DbSet, model config and table (migration DropDriverHostStatusTable — empty on every deployment, so the scaffolder's data-loss warning is moot, and Down() recreates it exactly). Found en route, NOT fixed here: `DriverInstanceResilienceStatus` is the identical defect — no writer, no reader, only a DbSet declaration, while the live data rides the `driver-resilience-status` telemetry channel. Its doc-comment now states that rather than describing the sampler and AdminUI join that were never built. Filed as #524 rather than widening this schema change beyond what was asked. Claude-Session: https://claude.ai/code/session_015p7wGqy3YpZNCpDzTpGMKo
This commit is contained in:
Joseph Doherty
@@ -1,131 +0,0 @@
|
||||
using Microsoft.EntityFrameworkCore;
|
||||
using Shouldly;
|
||||
using Xunit;
|
||||
using ZB.MOM.WW.OtOpcUa.Configuration.Entities;
|
||||
using ZB.MOM.WW.OtOpcUa.Configuration.Enums;
|
||||
|
||||
namespace ZB.MOM.WW.OtOpcUa.Configuration.Tests;
|
||||
|
||||
/// <summary>
|
||||
/// End-to-end round-trip through the DB for the <see cref="DriverHostStatus"/> entity
|
||||
/// added in PR 33 — exercises the composite primary key (NodeId, DriverInstanceId,
|
||||
/// HostName), string-backed <c>DriverHostState</c> conversion, and the two indexes the
|
||||
/// Admin UI's drill-down queries will scan (NodeId, LastSeenUtc).
|
||||
/// </summary>
|
||||
[Trait("Category", "SchemaCompliance")]
|
||||
[Collection(nameof(SchemaComplianceCollection))]
|
||||
public sealed class DriverHostStatusTests(SchemaComplianceFixture fixture)
|
||||
{
|
||||
/// <summary>Verifies that the composite key allows the same host across different nodes or drivers.</summary>
|
||||
[Fact]
|
||||
public async Task Composite_key_allows_same_host_across_different_nodes_or_drivers()
|
||||
{
|
||||
await using var ctx = NewContext();
|
||||
|
||||
// Same HostName + DriverInstanceId across two different server nodes — classic 2-node
|
||||
// redundancy case. Both rows must be insertable because each server node owns its own
|
||||
// runtime view of the shared host.
|
||||
var now = DateTime.UtcNow;
|
||||
ctx.DriverHostStatuses.Add(new DriverHostStatus
|
||||
{
|
||||
NodeId = "node-a", DriverInstanceId = "galaxy-1", HostName = "GRPlatform",
|
||||
State = DriverHostState.Running,
|
||||
StateChangedUtc = now, LastSeenUtc = now,
|
||||
});
|
||||
ctx.DriverHostStatuses.Add(new DriverHostStatus
|
||||
{
|
||||
NodeId = "node-b", DriverInstanceId = "galaxy-1", HostName = "GRPlatform",
|
||||
State = DriverHostState.Stopped,
|
||||
StateChangedUtc = now, LastSeenUtc = now,
|
||||
Detail = "secondary hasn't taken over yet",
|
||||
});
|
||||
// Same server node + host, different driver instance — second driver doesn't clobber.
|
||||
ctx.DriverHostStatuses.Add(new DriverHostStatus
|
||||
{
|
||||
NodeId = "node-a", DriverInstanceId = "modbus-plc1", HostName = "GRPlatform",
|
||||
State = DriverHostState.Running,
|
||||
StateChangedUtc = now, LastSeenUtc = now,
|
||||
});
|
||||
await ctx.SaveChangesAsync();
|
||||
|
||||
var rows = await ctx.DriverHostStatuses.AsNoTracking()
|
||||
.Where(r => r.HostName == "GRPlatform").ToListAsync();
|
||||
|
||||
rows.Count.ShouldBe(3);
|
||||
rows.ShouldContain(r => r.NodeId == "node-a" && r.DriverInstanceId == "galaxy-1");
|
||||
rows.ShouldContain(r => r.NodeId == "node-b" && r.State == DriverHostState.Stopped && r.Detail == "secondary hasn't taken over yet");
|
||||
rows.ShouldContain(r => r.NodeId == "node-a" && r.DriverInstanceId == "modbus-plc1");
|
||||
}
|
||||
|
||||
/// <summary>Verifies that the upsert pattern updates existing records in place.</summary>
|
||||
[Fact]
|
||||
public async Task Upsert_pattern_for_same_key_updates_in_place()
|
||||
{
|
||||
// The publisher hosted service (follow-up PR) upserts on every transition +
|
||||
// heartbeat. This test pins the two-step pattern it will use: check-then-add-or-update
|
||||
// keyed on the composite PK. If the composite key ever changes, this test breaks
|
||||
// loudly so the publisher gets a synchronized update.
|
||||
await using var ctx = NewContext();
|
||||
var t0 = DateTime.UtcNow;
|
||||
ctx.DriverHostStatuses.Add(new DriverHostStatus
|
||||
{
|
||||
NodeId = "upsert-node", DriverInstanceId = "upsert-driver", HostName = "upsert-host",
|
||||
State = DriverHostState.Running,
|
||||
StateChangedUtc = t0, LastSeenUtc = t0,
|
||||
});
|
||||
await ctx.SaveChangesAsync();
|
||||
|
||||
var t1 = t0.AddSeconds(30);
|
||||
await using (var ctx2 = NewContext())
|
||||
{
|
||||
var existing = await ctx2.DriverHostStatuses.SingleAsync(r =>
|
||||
r.NodeId == "upsert-node" && r.DriverInstanceId == "upsert-driver" && r.HostName == "upsert-host");
|
||||
existing.State = DriverHostState.Faulted;
|
||||
existing.StateChangedUtc = t1;
|
||||
existing.LastSeenUtc = t1;
|
||||
existing.Detail = "transport reset by peer";
|
||||
await ctx2.SaveChangesAsync();
|
||||
}
|
||||
|
||||
await using var ctx3 = NewContext();
|
||||
var final = await ctx3.DriverHostStatuses.AsNoTracking().SingleAsync(r =>
|
||||
r.NodeId == "upsert-node" && r.HostName == "upsert-host");
|
||||
final.State.ShouldBe(DriverHostState.Faulted);
|
||||
final.Detail.ShouldBe("transport reset by peer");
|
||||
// Only one row — a naive "always insert" would have created a duplicate PK and thrown.
|
||||
(await ctx3.DriverHostStatuses.CountAsync(r => r.NodeId == "upsert-node")).ShouldBe(1);
|
||||
}
|
||||
|
||||
/// <summary>Verifies that the State enum is persisted as a string, not an integer.</summary>
|
||||
[Fact]
|
||||
public async Task Enum_persists_as_string_not_int()
|
||||
{
|
||||
// Fluent config sets HasConversion<string>() on State — the DB stores 'Running' /
|
||||
// 'Stopped' / 'Faulted' / 'Unknown' as nvarchar(16). Verify by reading the raw
|
||||
// string back via ADO; if someone drops the conversion the column will contain '1'
|
||||
// / '2' / '3' and this assertion fails. Matters because DBAs inspecting the table
|
||||
// directly should see readable state names, not enum ordinals.
|
||||
await using var ctx = NewContext();
|
||||
ctx.DriverHostStatuses.Add(new DriverHostStatus
|
||||
{
|
||||
NodeId = "enum-node", DriverInstanceId = "enum-driver", HostName = "enum-host",
|
||||
State = DriverHostState.Faulted,
|
||||
StateChangedUtc = DateTime.UtcNow, LastSeenUtc = DateTime.UtcNow,
|
||||
});
|
||||
await ctx.SaveChangesAsync();
|
||||
|
||||
await using var conn = fixture.OpenConnection();
|
||||
using var cmd = conn.CreateCommand();
|
||||
cmd.CommandText = "SELECT [State] FROM DriverHostStatus WHERE NodeId = 'enum-node'";
|
||||
var rawValue = (string?)await cmd.ExecuteScalarAsync();
|
||||
rawValue.ShouldBe("Faulted");
|
||||
}
|
||||
|
||||
private OtOpcUaConfigDbContext NewContext()
|
||||
{
|
||||
var options = new DbContextOptionsBuilder<OtOpcUaConfigDbContext>()
|
||||
.UseSqlServer(fixture.ConnectionString)
|
||||
.Options;
|
||||
return new OtOpcUaConfigDbContext(options);
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user